JP2017211430A - Information processing device and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device and an information processing method with which it is possible to correct texts via speech input.SOLUTION: The information processing device comprises a transmitting unit for transmitting speech information including a text correction command and a correction target, and a receiving unit for receiving processing results based on the correction command and the correction target.SELECTED DRAWING: Figure 1

Description

  The present disclosure relates to an information processing apparatus and an information processing method.

  In recent years, voice command input technology has been developed. In voice command input, for example, a speech recognition system recognizes a user utterance as text, performs syntax analysis of the recognized text, and executes a command according to the analysis result. Regarding such a speech recognition system, for example, Patent Document 1 below describes a speech recognition correction method for correcting a speech recognition result using context information. The context information includes a user input history and a conversation history.

JP2015-018265A

  However, when characters are input by voice, deletion or correction of characters, switching of the type of characters to be input, etc. require operation from a physical character input interface, or deletion or correction is performed by voice. And the text is input as it is as a voice recognition result.

  Therefore, the present disclosure proposes an information processing apparatus and an information processing method capable of realizing sentence proofreading by voice input.

  According to the present disclosure, an information processing apparatus is provided that includes: a transmission unit that transmits audio information including a sentence calibration command and a calibration target; and a reception unit that receives a processing result based on the calibration command and the calibration target. .

  According to the present disclosure, an information processing apparatus is provided that includes a receiving unit that receives audio information including a sentence correction command and a correction target, and a transmission unit that transmits a processing result based on the correction command and the correction target. .

  According to the present disclosure, a processor proposes an information processing method including: transmitting a voice information including a sentence proofreading instruction and a proofreading target; and receiving an analysis result based on the proofreading instruction and the proofreading target. To do.

  According to the present disclosure, a processor proposes an information processing method including receiving audio information including a sentence proofreading instruction and a proofreading target, and transmitting an analysis result based on the proofreading instruction and the proofreading target. To do.

  As described above, according to the present disclosure, it is possible to realize sentence proofreading by voice input.

  Note that the above effects are not necessarily limited, and any of the effects shown in the present specification, or other effects that can be grasped from the present specification, together with or in place of the above effects. May be played.

It is a figure explaining the outline | summary of the information processing system by this embodiment. It is a block diagram which shows an example of a structure of the client terminal by this embodiment. It is a block diagram which shows an example of a structure of the server by this embodiment. It is a figure which shows the specific example in the case of performing designation | designated of the kind of character to input by this embodiment with a sound. It is a figure which shows the specific example in the case of performing specification of the kanji conversion of the character to input by this embodiment by a voice. It is a figure which shows an example of the user utterance by this embodiment, and the analysis result of proofreading information. It is a figure which shows an example of the final output result with respect to the user utterance shown in FIG. It is a figure which shows an example of the analysis result of the proofreading information which considered the user utterance and context information by this embodiment. It is a figure which shows an example of the final output result with respect to the user utterance shown in FIG. It is a flowchart which shows the operation processing of the information processing system by this embodiment. It is a figure which shows the other system structure by this embodiment. It is a block diagram which shows an example of a structure of the edge server by this embodiment.

  Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

The description will be made in the following order.
1. 1. Overview of information processing system according to an embodiment of the present disclosure Configuration 2-1. Configuration of client terminal 2-2. 2. Server configuration Operation processing 4. Other system configuration Summary

<< 1. Overview of Information Processing System According to One Embodiment of Present Disclosure >>
First, an overview of an information processing system according to an embodiment of the present disclosure will be described. FIG. 1 is a diagram for explaining the outline of the information processing system according to this embodiment. As shown in FIG. 1, the information processing system according to the present embodiment includes a client terminal 1 and a server 2. The client terminal 1 and the server 2 are connected via, for example, the network 3 and transmit / receive data.

  The information processing system according to the present embodiment is a speech recognition system that realizes character input by speech, and performs speech recognition and text analysis of user utterances collected by the client terminal 1, and text is sent to the client terminal 1 as an analysis result. Output.

  The client terminal 1 may be, for example, a smartphone, a tablet terminal, a mobile phone terminal, a wearable terminal, a personal computer, a game machine, a music player, or the like.

  Here, in the existing speech recognition system, it is difficult to switch the character type (switching between uppercase, lowercase, Roman, numbers, hiragana, katakana, etc.) by voice, and operation from a physical character input interface is difficult. Was necessary. Further, when the input sentence is proofread, if the input of deletion, insertion, correction or the like of the sentence is made by voice, the character is inputted as it is as a voice recognition result, so that the proofreading by voice is difficult.

  In addition, because kanji has homonyms, the target kanji cannot be obtained with a single conversion, or the kanji desired by the user cannot be produced, so it has been necessary to switch to a physical character input interface.

  Therefore, the information processing system according to the present embodiment realizes sentence proofreading by voice input and eliminates a complicated operation such as switching to a physical character input interface at the time of proofreading. Specifically, the information processing system according to the present embodiment determines whether the utterance is a proof utterance or a normal utterance in the text analysis of the user utterance, and analyzes the proofreading information in the case of the proof utterance.

  The overview of the information processing system according to the present embodiment has been described above. Next, the configuration of each device included in the information processing system according to the present embodiment will be described with reference to FIGS.

<< 2. Configuration >>
<2-1. Configuration of client terminal>
FIG. 2 is a block diagram illustrating an example of the configuration of the client terminal 1 according to the present embodiment. As illustrated in FIG. 2, the client terminal 1 (information processing apparatus) includes a control unit 10, a voice input unit 11, an imaging unit 12, a sensor 13, a communication unit 14, a display unit 15, and a storage unit 16.

  The control unit 10 functions as an arithmetic processing device and a control device, and controls the overall operation in the client terminal 1 according to various programs. The control unit 10 is realized by an electronic circuit such as a CPU (Central Processing Unit) or a microprocessor, for example. The control unit 10 may include a ROM (Read Only Memory) that stores programs to be used, calculation parameters, and the like, and a RAM (Random Access Memory) that temporarily stores parameters that change as appropriate.

  The control unit 10 transmits the voice of the user utterance input from the voice input unit 11 from the communication unit 14 to the server 2 via the network 3. The form of voice information to be transmitted may be collected voice data (raw data) or feature amount data extracted from the collected voice data (processed to some extent, such as phoneme string). Alternatively, it may be a text analysis result of the collected voice data. The text analysis result of the speech data is a result of analyzing a calibration command portion and a calibration target portion included in the speech of the user utterance, for example. Such an analysis can be performed by the local text analysis unit 102 described below. In the present specification, the “calibration command” indicates what kind of calibration should be performed on the calibration target, for example, correction of an input character string such as deletion, replacement, addition, It is assumed that the character type to be input (alphabet, uppercase, lowercase, hiragana, katakana, etc.) and the expression specification of the input character (kanji, spelling, etc.) are assumed. Further, in this specification, “calibration target” indicates a target for a calibration command.

  Also, the control unit 10 uses the captured image of the user action captured by the imaging unit 12 and the sensor information (such as touch information on the screen) detected by the sensor 13 during user utterance as context information from the communication unit 14 to the network. 3 to the server 2. The form of the context information to be transmitted may be an acquired captured image or sensor information (raw data), or feature amount data extracted from the acquired captured image or sensor information (processed to some extent such as vectorization). It may be an acquired captured image or an analysis result (recognition result) of sensor information. The analysis result of the captured image and sensor information is, for example, the result of recognizing the user's operation and operation.

  As shown in FIG. 2, the control unit 10 can also function as a local speech recognition unit 101, a local text analysis unit 102, and a local final output determination unit 103.

  The local speech recognition unit 101 performs speech recognition on the speech signal of the user utterance input from the speech input unit 11, and converts the user utterance into text. The local speech recognition unit 101 according to the present embodiment is a subset of the speech recognition unit 201 of the server 2 described later, and has a simple speech recognition function.

  The local text analysis unit 102 analyzes a character string converted into text by voice recognition. Specifically, the local text analysis unit 102 refers to the proof utterance data stored in advance in the storage unit 16 and analyzes whether the character string is a mere utterance of character input (normal utterance) or a proof utterance. To do. The local text analysis unit 102 outputs the proofreading utterance likelihood and the proofreading target and the proofreading command in the case of the proofreading utterance. The likelihood of proofreading utterance is calculated as a score indicating confidence. Further, the local text analysis unit 102 may output a plurality of candidates together with the score. Further, the local text analysis unit 102 may analyze the captured image captured by the imaging unit 12 during user utterance and other sensor information (acceleration sensor information, touch sensor information, etc.) detected by the sensor 13. Good. The local text analysis unit 102 according to the present embodiment is a subset of the text analysis unit 202 of the server 2 to be described later and has a simple analysis function. Specifically, since the amount of proofreading utterance data used in the local text analysis unit 102 is smaller than the amount of data held by the server 2, for example, the proofreading term “deletion” can be understood, but “desired” “ I don't understand words like "I want you to erase them" as proofreading terms.

  The local final output determination unit 103 has a function of determining what is finally output. For example, the local final output determination unit 103 determines whether the user utterance is a normal utterance or a proofreading utterance based on a specific keyword (for example, “proofreading mode”, “switching”, etc.) extracted by voice recognition and a text analysis result. . If it is determined that the utterance is normal, the local final output determination unit 103 outputs the speech-recognized character string as it is on the screen of the display unit 15. On the other hand, when it is determined that the utterance is proofreading, the local final output determination unit 103 performs a proofreading process on the input sentence based on the proofreading target and the proofreading command analyzed by the local text analysis unit 102, and displays the proofreading result. 15 screens are output. When there are a plurality of analysis results, the local final output determination unit 103 may determine which analysis result to use with reference to a score indicating the certainty factor of each candidate.

  Note that the local final output determination unit 103 according to the present embodiment is a subset of the final output determination unit 203 of the server 2 to be described later, and has a simple determination function.

  The functional configuration of the control unit 10 has been described above. The control unit 10 can increase the processing speed by performing processing in the local subsets such as the local speech recognition unit 101, the local text analysis unit 102, and the local final output determination unit 103. It is not limited. For example, the control unit 10 can send data to the server 2 to request processing when a subset cannot perform sufficient processing or an error occurs, and can receive and use the processing result from the server 2. Good. Alternatively, the control unit 10 transmits data to the server 2 to request processing, and also performs processing in the subset, waits for a predetermined time for the processing result from the server 2, or refers to a score indicating the certainty of each processing result Then, data to be used may be selected.

  The voice input unit 11 collects user voice and surrounding environmental sounds, and outputs a voice signal to the control unit 10. Specifically, the voice input unit 11 is realized by a microphone, an amplifier, and the like. The audio input unit 11 may be realized by a microphone array including a plurality of microphones.

  The imaging unit 12 captures the user's face periphery and motion, and outputs the captured image to the control unit 10. The imaging unit 12 includes a lens system including an imaging lens, a diaphragm, a zoom lens, and a focus lens, a drive system that causes the lens system to perform a focus operation and a zoom operation, and imaging light obtained by the lens system. And a solid-state imaging device array that generates an imaging signal through photoelectric conversion. The solid-state imaging device array may be realized by, for example, a CCD (Charge Coupled Device) sensor array or a CMOS (Complementary Metal Oxide Semiconductor) sensor array.

  The sensor 13 is a generic name for various sensors other than the imaging unit 12 (imaging sensor). For example, an acceleration sensor, a gyro sensor, a touch sensor provided on the screen of the display unit 15, and the like are assumed. The sensor 13 outputs the detected sensor information to the control unit 10.

  The communication unit 14 is a communication module that transmits and receives data to and from other devices by wire / wireless. The communication unit 14 is, for example, a wired LAN (Local Area Network), wireless LAN, Wi-Fi (Wireless Fidelity (registered trademark)), infrared communication, Bluetooth (registered trademark), short-range / non-contact communication, etc. Communicate directly or via a network access point.

  The display unit 15 is realized by, for example, a liquid crystal display (LCD) device, an OLED (Organic Light Emitting Diode) device, or the like. The display unit 15 displays information on the display screen according to the control of the control unit 10.

  The storage unit 16 stores a program or the like for the control unit 10 to execute various processes. The storage unit 16 includes a storage device including a storage medium, a recording device that records data on the storage medium, a reading device that reads data from the storage medium, and a deletion device that deletes data recorded on the storage medium. .

  The configuration of the client terminal 1 according to the present embodiment has been specifically described above. Note that the configuration of the client terminal 1 according to the present embodiment is not limited to the example shown in FIG. For example, the client terminal 1 may be configured not to include all or part of the local speech recognition unit 101, the local text analysis unit 102, and the local final output determination unit 103.

  In the present embodiment, the information processing system including the client terminal 1 and the server 2 is described. However, a single information processing apparatus having each configuration described with reference to FIGS. 2 to 3 as a client module and a server module. May be realized. Alternatively, the configuration of the client terminal 1 has the same function as each configuration (speech recognition unit 201, text analysis unit 202, final output determination unit 203) of the control unit 20 of the server 2 described with reference to FIG. It is good.

<2-2. Server configuration>
FIG. 3 is a block diagram illustrating an example of the configuration of the server 2 according to the present embodiment. As illustrated in FIG. 3, the server 2 (information processing apparatus) includes a control unit 20, a communication unit 21, and a proof speech DB (database) 22.

  The control unit 20 functions as an arithmetic processing device and a control device, and controls the overall operation in the server 2 according to various programs. The control unit 20 is realized by an electronic circuit such as a CPU (Central Processing Unit) and a microprocessor, for example. The control unit 20 may include a ROM (Read Only Memory) that stores programs to be used, calculation parameters, and the like, and a RAM (Random Access Memory) that temporarily stores parameters that change as appropriate.

  The control unit 20 performs a speech recognition process, a text analysis process, and a final output determination process based on the voice of the user utterance received from the client terminal 1, and processes the result (speech recognition result, text analysis result, or calibration information (for example, Control is performed to transmit the calibration result)) to the client terminal 1.

  As illustrated in FIG. 3, the control unit 20 can also function as a voice recognition unit 201, a text analysis unit 202, and a final output determination unit 203.

  The voice recognition unit 201 performs voice recognition on the voice signal of the user utterance transmitted from the client terminal 1 and converts the user utterance into text.

  The text analysis unit 202 analyzes a character string converted into text by voice recognition. Specifically, the text analysis unit 202 refers to the proofreading utterance data stored in advance in the proofreading utterance DB 22 and analyzes whether the character string is a simple character input utterance (normal utterance) or a proofreading utterance. . The text analysis unit 202 outputs the proofreading utterance likelihood, the proofreading target and the proofreading command in the case of the proofreading utterance. The likelihood of proofreading utterance is calculated as a score indicating confidence. The text analysis unit 202 may output a plurality of candidates together with the score. Further, the text analysis unit 202 may perform analysis in consideration of context information (captured image and sensor information) at the time of user utterance transmitted from the client terminal 1.

  The analysis of the proofreading information is not limited to the method using the proofreading utterance DB 22 generated in advance. For example, the accuracy of the proofreading information analysis can be increased by using machine learning.

  The final output determination unit 203 has a function of determining what is finally output. For example, the final output determination unit 203 determines whether the user utterance is a normal utterance or a proofreading utterance based on a specific keyword (for example, “proofreading mode” “switching”, etc.) extracted by voice recognition or a text analysis result. When there are a plurality of analysis results, the final output determination unit 203 may determine which analysis result to use with reference to a score indicating the certainty factor of each candidate.

  When it is determined that the utterance is normal, the final output determination unit 203 transmits the speech-recognized character string from the communication unit 21 to the client terminal 1. On the other hand, when it is determined that the utterance is proofreading, the final output determining unit 203 analyzes the proofreading target based on the proofreading command analyzed by the text analyzing unit 202 and finally determines the proofreading result as proofreading information from the communication unit 21 to the client terminal. 1 to send.

  Further, the final output determination unit 203 analyzes the captured image obtained by capturing the user's action by the imaging unit 12 transmitted from the client terminal 1 as the context information, detects the body movement registered in advance, Switching between the normal input mode and the sentence proofreading mode may be performed. Alternatively, the final output determination unit 203 analyzes the sensor information detected by the sensor 13 transmitted from the client terminal 1 as the context information, and moves in advance (for example, shakes the screen, touches the screen, etc.) ) May be detected to switch between the normal input mode and the sentence proofreading mode.

  Further, the final output determination unit 203 can determine whether the utterance is a proofreading speech by combining the text analysis result of the user utterance with the captured image and sensor information. For example, when the user utters “delete all from here” while showing the characters displayed on the screen, the final output determination unit 203 determines from the analysis result of the utterance contents and the operation indicating the characters on the screen. , Judge that it is a sentence proofreading mode.

  Here, a user utterance example and a final output example of each utterance according to the present embodiment will be specifically described with reference to FIGS.

(A) Designation of Character Type FIG. 4 is a diagram showing a specific example in the case of designating the type of character to be input by voice. For example, as shown in the first line of FIG. 4, when the user utterance is “Katakana no Kyoto”, the voice recognition unit 201 outputs a character string such as “Tokyo Tower of Katakana” by voice recognition. In this case, the existing voice recognition system may output “Katakana no Tokyo Tower” without changing the voice-recognized character string. On the other hand, in this embodiment, text analysis is performed with reference to the proofreading utterance data with respect to the voice-recognized character string, `` Katakana '' is analyzed from the voice recognition result as the proof designation of the character type `` Katakana '', and `` Analyze Tokyo Tower as a calibration target. As a result, as shown in the first line of FIG. 4, the final output result is “Tokyo Tower” expressed in katakana.

  Also, as shown in the second line of FIG. 4, when the user utterance is “Emoji can only be ignored”, the voice recognition unit 201 outputs a character string such as “Michael with only M” by voice recognition. . In this case, in the existing speech recognition system, there is a possibility that the character string that has been speech-recognized is output as “Michael only in M”. On the other hand, in this embodiment, text analysis is performed with reference to the proofreading utterance data with respect to the voice-recognized character string, and “M only uppercase” is designated as the character type from the voice recognition result. And analyze "Michael" as a calibration target. As a result, the final output result is “Michael” as shown in the second line of FIG.

(B) Utilization of Sound and Transcription FIG. 5 is a diagram showing a specific example in the case where designation of Kanji conversion of input characters is performed by voice. For example, as shown in the first line of FIG. 5, when the user utterance is “Yukyu Kyu no Yuuni Kodomo no Moko”, the voice recognition unit 201 uses a character recognition such as “Children with a paid vacation” by voice recognition. Output a column. In this case, the existing voice recognition system may output “children with paid leave” without changing the voice-recognized character string. On the other hand, in the present embodiment, text analysis is performed with reference to the proofreading utterance data for the speech-recognized character string, “paid leave” is analyzed from the speech recognition result as the proofreading designation of the kanji, and “yes” is set. Analyze with calibration target. In addition, “child” is analyzed as a proofreading designation of kanji, and “child” is analyzed as a proofreading target. As a result, as shown in the first line of FIG. 5, the final output result becomes “child” expressed in kanji desired by the user. Even if there are other kanji candidates corresponding to the sound “Yuko”, it is possible to input in the kanji desired by the user.

  As shown in the second line of FIG. 5, when the user utterance is “Shiratori Totori is Tottori Tori”, the speech recognition unit 201 uses the speech recognition to read “Swan Bird is Tottori Tori”. Output a column. In this case, the existing speech recognition system may output “swan bird is Tottori-tori” as it is as the speech-recognized character string. On the other hand, in the present embodiment, text analysis is performed with reference to the proofreading utterance data with respect to the voice-recognized character string, and “Swan Bird is Tottori-no-Tori” is analyzed from the voice recognition result as proofreading designation of kanji. Analyze “Swan” as a calibration target. As a result, as shown in the second line of FIG. 5, the final output result is “white picked” expressed in the user-desired kanji. Even when there are other kanji candidates corresponding to the sound “Shiratori”, it is possible to input in kanji desired by the user.

(C) Calibration location and operation instruction It is also possible to instruct the calibration target range and calibration contents by voice. For example, an example of the analysis result of the user utterance and the proofreading information as shown below is given.

  Furthermore, an example is demonstrated with reference to FIG. 6 and FIG. FIG. 6 is a diagram illustrating an example of a user utterance and calibration information analysis result according to the present embodiment. As shown in FIG. 6, when the user utterance is “all the things that have been done from the location,” the speech recognition unit 201 uses the speech recognition to indicate “ The character string such as “Continue to consider by deleting all the characters from the bottom” is output. In this case, in the existing speech recognition system, there is a risk that the text string that has been speech-recognized may be output as “Continue to consider all the following items from the bottom”. On the other hand, in the present embodiment, text analysis is performed with reference to the proofreading utterance data for the speech-recognized character string, and from the speech recognition result, “proofreading designation: corrected to“ continuous examination ””, “calibration target:“ following Analyze "Proposal" and later ".

  FIG. 7 is a diagram illustrating an example of a final output result for the user utterance illustrated in FIG. 6. As shown in FIG. 7, in the input text displayed on the screen 30, a screen 31 in which “the following cases” and later are deleted and corrected to “continuous examination” is output as a final output result.

(D) Utilization of Context Information Next, an example of a calibration process considering the context information will be described. In the present embodiment, it is possible to perform a text analysis and a calibration analysis in consideration of a captured image and sensor information acquired at the time of user utterance.

  Here, an example using sensor information detected by a touch sensor provided in the display unit 15 will be described with reference to FIGS. 8 and 9. FIG. 8 is a diagram illustrating an example of the analysis result of the calibration information in consideration of the user utterance and the context information according to the present embodiment. As shown in FIG. 8, when the user utterance is “please make it here”, the speech recognition unit 201 outputs a character string “make here in the morning” by speech recognition. Further, sensor information indicating the position coordinates (x, y) on the screen detected by the touch sensor of the display unit 15 at the time of user utterance is acquired.

  In this case, in the existing speech recognition system, there is a possibility that the text string that has been speech-recognized is output as “here in the morning”. On the other hand, in the present embodiment, text analysis is performed with reference to the calibration utterance data and touch sensor information with respect to the voice-recognized character string, and “calibration designation: amended to“ AM ””, “calibration target: coordinates (x, y) ”.

  FIG. 8 is a diagram showing an example of a final output result for the user utterance shown in FIG. As shown in FIG. 8, in the input text displayed on the screen 32, the character “PM” corresponding to the coordinates (x, y) touched by the user is deleted and corrected to “AM”. 33 is output as the final output result.

  In the above-described example, the coordinate position on the screen is detected by the touch sensor, but the present embodiment is not limited to this, and can be similarly realized if the user's line of sight can be accurately captured. That is, for example, the position on the screen on which the user is gazing at the time of the user utterance of “Make this here in the morning” is detected by the gaze sensor (gaze tracker) and considered as context information.

Further, if a point of interest, a range, or an area on the screen can be specified by the user's line of sight, it is possible to automatically narrow down the user's desired candidates from a plurality of candidate options displayed on the screen.

  In this embodiment, when a position on the screen is designated as “here” or “this side”, the background color of the character string portion corresponding to the coordinates (x, y) is changed, and the like is fed back to the user. Then, it is possible to confirm the attention location and range. The user can verbally answer such as “OK there” or “No”.

(E) Use of Keywords Next, an example of a proofreading process when a specific keyword is extracted from a speech uttered user utterance will be described. When the user utterance is “A, as in Adam. D, as in Denver. T, as in Thomas.”, The speech recognition unit 201 performs “A, as in Adam. D, as in Denver. T, as in Thomas. " In this case, the existing speech recognition system may output "A, as in Adam. D, as in Denver. T, as in Thomas." On the other hand, in the present embodiment, text analysis is performed with reference to proofreading utterance data for a speech-recognized character string, and the spelling of the alphabet such as “Adam” “Denver” “Thomas” is transmitted from the speech recognition result. When the keyword to be used is extracted, it is analyzed as “Calibration designation: alphabet” and “Calibration target:“ A ”“ D ”“ T ””. As a result, the final output result is “ADT” expressed in the spelling desired by the user.

  The communication unit 21 is connected to an external device and transmits / receives data. For example, the communication unit 21 receives voice information and context information of a user utterance from the client terminal 1, transmits the above-described voice recognition processing result, text analysis processing result, or final output determination processing result to the client terminal 1. To do.

  The calibration utterance DB 22 is a storage unit that stores calibration utterance data collected in large quantities in advance, and includes a storage medium, a recording device that records data in the storage medium, a reading device that reads data from the storage medium, and a recording medium. The storage device includes a deletion device that deletes the recorded data. The proofreading utterance data includes, for example, keywords and sentence examples used for proofreading utterances.

<< 3. Action processing >>
Subsequently, an operation process of the information processing system according to the present embodiment will be described with reference to FIG. FIG. 10 is a flowchart showing an operation process of the information processing system according to the present embodiment. The following process may be performed by at least one of the control unit 10 of the client terminal 1 and the control unit 20 of the server 2.

  As shown in FIG. 10, first, a user utterance (voice information) is acquired (step S100), and voice recognition is performed on the user utterance (step S103).

  Next, text analysis is performed on the character string output by voice recognition (step S106). Specifically, referring to the proofreading utterance data, the likelihood of the proofreading of the character string and the proofreading information in the case of the proofreading utterance are analyzed. Context information acquired at the time of user utterance may be used.

  Next, a final output is determined based on the text analysis result (step S109). Also in this case, the context information acquired at the time of user utterance may be used.

  Next, when it is determined that the normal utterance is determined by the final output determination, the character string of the voice recognition result is output as it is (step S112).

  On the other hand, if it is determined that the utterance is proofread based on the final output determination, the sentence is proofread and the proofreading result is output (step S115).

  The operation processing of the information processing system according to the present embodiment has been described above.

<< 4. Other system configuration >>
The configuration of the information processing system according to the present embodiment is not limited to the example illustrated in FIG. 1. For example, as illustrated in FIG. 11, a system configuration including an edge server 4 that enables processing distribution may be used. FIG. 11 is a diagram showing another system configuration according to this embodiment. As shown in FIG. 11, another system configuration including a client terminal 1, a server 2, and an edge server 4 is conceivable.

  A configuration example of the edge server 4 according to the present embodiment is shown in FIG. As illustrated in FIG. 12, the edge server 4 includes a control unit 40, a communication unit 41, and an edge side calibration utterance DB 42. The control unit 40 also functions as an edge side speech recognition unit 401, an edge side text analysis unit 402, and an edge side final output determination unit 403. The edge-side speech recognition unit 401 is a subset of the speech recognition unit 201 of the server 2 (hereinafter referred to as an external subset), and the edge-side text analysis unit 402 is an external subset of the text analysis unit 202 and includes an edge side. The final output determination unit 403 is an external subset of the final output determination unit 203.

  The edge server 4 is a medium-scale processing server as compared with the server 2, but is disposed near the client terminal 1 in terms of communication distance, and can be more accurate than the client terminal 1 and reduce communication delay. It is.

  The client terminal 1 sends data to the edge server 4 when a sufficient processing cannot be performed with its own subset or when an error occurs, and receives a processing result from the edge server 4 for use. May be. Alternatively, the client terminal 1 transmits data to the edge server 4 and the server 2 to make a processing request, and also performs processing on a subset that the client terminal 1 has, and waits for a predetermined time from the processing results from the edge server 4 and the server 2, Data to be used may be selected with reference to a score indicating the certainty factor of each processing result.

<< 5. Summary >>
As described above, according to the information processing system of the present embodiment, it is possible to realize sentence proofreading by voice input.

  The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the present technology is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field of the present disclosure can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that it belongs to the technical scope of the present disclosure.

  For example, it is possible to create a computer program for causing the client terminal 1 or the server 2 to exhibit the functions of the client terminal 1 or the server 2 on hardware such as the CPU, ROM, and RAM incorporated in the client terminal 1 or the server 2 described above. A computer-readable storage medium storing the computer program is also provided.

  Further, the effects described in the present specification are merely illustrative or exemplary and are not limited. That is, the technology according to the present disclosure can exhibit other effects that are apparent to those skilled in the art from the description of the present specification in addition to or instead of the above effects.

In addition, this technique can also take the following structures.
(1)
A transmitter for transmitting voice information including a sentence proofreading command and a proofreading target;
A receiving unit for receiving a processing result based on the calibration command and the calibration target;
An information processing apparatus comprising:
(2)
The information processing apparatus according to (1), wherein the voice information is collected user voice data.
(3)
The information processing apparatus according to (1), wherein the voice information is feature amount data extracted from collected user voice data.
(4)
The information processing apparatus according to (1), wherein the voice information is data indicating a calibration command and a calibration target recognized from collected user voice data.
(5)
The transmission unit transmits context information at the time of voice input together with the voice information,
The information processing apparatus according to any one of (1) to (4), wherein the reception unit receives a processing result based on the calibration command, a calibration target, and the context information.
(6)
The information processing apparatus according to (5), wherein the context information is sensor information that detects a user action.
(7)
The information processing apparatus according to (5), wherein the context information is feature amount data extracted from sensor information that detects a user's operation.
(8)
The information processing apparatus according to (5), wherein the context information is data indicating a result recognized from sensor information obtained by detecting a user operation.
(9)
The processing result received by the receiving unit includes at least one of a speech recognition result of the transmitted speech information, a text analysis result, or calibration information based on a calibration command and a calibration target included in the speech information. The information processing apparatus according to any one of (8) to (8).
(10)
The information processing apparatus according to (9), wherein the processing result includes data indicating a certainty factor of the processing result.
(11)
The information processing apparatus according to (9) or (10), wherein the calibration information includes a calibration result obtained by processing a calibration target based on a calibration command that is finally determined.
(12)
A receiver for receiving voice information including a proofreading command and a proofreading target;
A transmitter for transmitting a processing result based on the calibration command and the calibration target;
An information processing apparatus comprising:
(13)
The processing result transmitted by the transmitting unit includes at least one of a speech recognition result of the received speech information, a text analysis result, or calibration information based on a calibration command and a calibration target included in the speech information. ).
(14)
The information processing apparatus according to (13), wherein the processing result includes data indicating a certainty factor of the processing result.
(15)
The information processing apparatus according to (13) or (14), wherein the calibration information includes a calibration result obtained by processing a calibration target based on a finally determined calibration command.
(16)
The receiving unit receives context information at the time of voice input together with the voice information,
The information processing apparatus according to any one of (12) to (15), wherein the transmission unit transmits a processing result based on the calibration command, a calibration target, and the context information.
(17)
Processor
Sending audio information including proofreading instructions and proofreading targets;
Receiving an analysis result based on the calibration command and the calibration target;
Including an information processing method.
(18)
Processor
Receiving audio information including proofreading instructions and proofreading targets;
Transmitting an analysis result based on the calibration command and the calibration target;
Including an information processing method.

DESCRIPTION OF SYMBOLS 1 Client terminal 10 Control part 101 Local speech recognition part 102 Local text analysis part 103 Local final output determination part 11 Voice input part 12 Imaging part 13 Sensor 14 Communication part 15 Display part 16 Storage part 2 Server 20 Control part 201 Voice recognition part 202 Text analysis unit 203 Final output determination unit 21 Communication unit 22 Proofreading DB
DESCRIPTION OF SYMBOLS 3 Network 4 Edge server 40 Control part 401 Edge side speech recognition part 402 Edge side text analysis part 403 Edge side final output determination part 41 Communication part 42 Edge side calibration utterance DB

Claims (18)

A transmitter for transmitting voice information including a sentence proofreading command and a proofreading target;
A receiving unit for receiving a processing result based on the calibration command and the calibration target;
An information processing apparatus comprising:   The information processing apparatus according to claim 1, wherein the voice information is collected user voice data.   The information processing apparatus according to claim 1, wherein the voice information is feature amount data extracted from collected user voice data.   The information processing apparatus according to claim 1, wherein the voice information is data indicating a calibration command and a calibration target recognized from collected user voice data. The transmission unit transmits context information at the time of voice input together with the voice information,
The information processing apparatus according to claim 1, wherein the reception unit receives a processing result based on the calibration command, a calibration target, and the context information.   The information processing apparatus according to claim 5, wherein the context information is sensor information that detects a user action.   The information processing apparatus according to claim 5, wherein the context information is feature amount data extracted from sensor information in which a user operation is detected.   The information processing apparatus according to claim 5, wherein the context information is data indicating a result recognized from sensor information detected by a user operation.   The processing result received by the receiving unit includes at least one of a speech recognition result of the transmitted speech information, a text analysis result, or calibration information based on a calibration command and a calibration target included in the speech information. The information processing apparatus described in 1.   The information processing apparatus according to claim 9, wherein the processing result includes data indicating a certainty factor of the processing result.   The information processing apparatus according to claim 9, wherein the calibration information includes a calibration result obtained by processing a calibration target based on a finally determined calibration command. A receiver for receiving voice information including a proofreading command and a proofreading target;
A transmitter for transmitting a processing result based on the calibration command and the calibration target;
An information processing apparatus comprising:   The processing result transmitted by the transmission unit includes at least one of a speech recognition result of the received speech information, a text analysis result, or calibration information based on a calibration command and a calibration target included in the speech information. The information processing apparatus described in 1.   The information processing apparatus according to claim 13, wherein the processing result includes data indicating a certainty factor of the processing result.   The information processing apparatus according to claim 13, wherein the calibration information includes a calibration result obtained by processing a calibration target based on a finally determined calibration command. The receiving unit receives context information at the time of voice input together with the voice information,
The information processing apparatus according to claim 12, wherein the transmission unit transmits a processing result based on the calibration command, a calibration target, and the context information. Processor
Sending audio information including proofreading instructions and proofreading targets;
Receiving an analysis result based on the calibration command and the calibration target;
Including an information processing method. Processor
Receiving audio information including proofreading instructions and proofreading targets;
Transmitting an analysis result based on the calibration command and the calibration target;
Including an information processing method.

Images